Water use in the São Francisco River basin, Brazil: an interregional input-output analysis

(1)

Water use in the São Francisco River basin, Brazil: an interregional

input-output analysis

Silveira, Suely F.R and Guilhoto, Joaquim José Martins

Universidade Federal de Viçosa, Universidade de São Paulo

1999

Online at https://mpra.ub.uni-muenchen.de/54674/

MPRA Paper No. 54674, posted 24 Mar 2014 11:59 UTC

(2)

Suely F.R. Silveira¹ and Joaquim J.M. Guilhoto²

Abstract

The planning and management of water resources aiming to a sustainable development must necessary pass through a series of studies that will reveal the interrelations and links among economic activities, and if one is considering that there is a set regions involved in the process, the direct and indirect regional interdependencies. Increasing demands of water-use for industrial, agricultural and urban sectors may became very competitive and can result in conflicts among multiple users. In the case of the São Francisco River Basin, Brazil, consisting of a wide-range five states - most of them having drought problems - the water in the basin plays an essential role and any activity in this area has to consider the effects of the water use in the intersectoral and interregional economic relations. Considering the three main states in the basin - Minas Gerais, Bahia and Pernambuco - an interregional input-output system for the economy and for the water flows was constructed by the authors. The above interregional system is then used to analyze the interregional and intersectoral dependencies among the states and the economic activities on the São Francisco river basin area and their relations with the use of water.

1 Introduction

As a resource, water use can be divided into 3 main uses: agriculture, industrial and domestic. Depending on the amount and quality of water available for use by the society it can be treated either as a free good or as market good.

The decision related to the water use in a region that comprises an hydrographic basin do depend on the relations among the states and the sectors that made use of this resource. So, the success of a economic policy may be very dependable of how a limited resource, like water, is used in the productive process.

1 Federal University of Viçosa (UFV), Brazil. E-mail: sfrsilve@carpa.ciagri.usp.br

2 University of São Paulo (USP), Brazil and University of Illinois. E-mail: guilhoto@usp.br

(3)

In the case of the São Francisco river basin that is locate in a region that has draught problems and comprises 5 Brazilian States, some of them with low level of development, the use of the water should be done such the most can be obtained from it.

Taking into consideration those aspects, this study tries to better understand how the water use takes place among the three most important states (Minas Gerais, Bahia, and Pernambuco), in terms of water use, in the São Francisco river base. To do so it is constructed an interregional input-model for the region and estimated the interactions that take place among the States in terms of production and water use.

In the next section it is presented a brief overview of the São Francisco River Basin. The third section will presented the methodology, in the forth section the results are presented, while some final remarks are made in the last section.

2 The São Francisco River Basin

It this study it is taking into consideration the states of Minas Gerais, Bahia and Pernambuco. As can be seen in Table 2-1 Minas Gerais has a share of 9% in the Brazilian GDP, Bahia 3% and Pernambuco 2%.

Table 2.1 - Main economical and geographical indicators of the Brazil, Minas Gerais, Bahia and Pernambuco.

Brazil Minas Gerais Bahia Pernambuco

Size (Km2) 8 511 996 588 383 561 026 98 281

Population (1996) 157 070 163 16 672 613 12 541 675 7 399 071 Urban 123 076 831 13 073 852 7 826 843 5 476 855

Rural 33 993 332 3 598 761 4 714 832 1 922 216

Urbanization (%) 78 78 62 74

GDP (1995) (US$ Million) 707 358 61 837 22 850 12 527

GDP per capita (US$) 4 554 3 737 1 801 1 678

Source: IBGE, 1997; Considera e Medina, 1998.

The São Francisco river basin occupies and area of 640 thousand km² (Figure 2.1), being 36,8% of it in Minas Gerais, 0,7% in Goiás and the remaining 62,5% in the states of Bahia,

(4)

Sergipe, Alagoas e Pernambuco. With an extension of 2700 km, there are around 14 million people living in the basin area (CODEVASF, 1998b).

From Table 2.2 one can see that in terms of area the States of Bahia, Minas Gerais, and Pernambuco are the ones more dependable from the São Francisco River basin.

Source: EMBRAPA - Sistema de Monitoramento por Satélite.

Figure 2.1 - São Francisco River Basin, Brazil.

(5)

Table 2.2 - Area and population of the states in the São Francisco river basin.

Area

(Km²) Population ⁽³⁾

1991 States Total Area Total

in the basin

(1)

Area in the basin ⁽²⁾

Total Urban Rural

Minas Gerais 588 383.6 262 201.9 235 471.3 6 931 099 5 667 175 1 263 924 Bahia 567 295.3 331 724.7 307 940.8 2 500 422 1 056 487 1 443 935 Distrito Federal 5 822.1 5 822.1 1 335.6 1 601 094 1 515 889 85 205

Goiás 341 289.5 13 133.8 3 141.8 94 383 71 494 22 889

Pernambuco⁽⁴⁾ 98 937.8 71 973.8 69 518.4 1 583 854 732 117 851 737

Sergipe 22 050.3 8 689.8 7 473.3 279 448 131 022 148 426

Alagoas 27 933.1 16 225.2 14 338.2 966 312 431 793 534 519

Total 1 651 711.7 709 771.3 639 219.4 13 956 612 9 605 977 4 350 635 Source: CODEVASF (1998b).

Notes: (1) Include total area of the borough partially in the Basin's area.

(2) Not include outside area of the borough partially in the Basin's area.

(3) Include total population of the borough partially in the Basin's area.

(4) Total area of the Pernambuco state include Fernando de Noronha's area.

3. Theoretical Background

3.1. The Rasmussen/Hirschman Approach

.U_i  B_i_ / /n B^* (2)

One of the criticisms of the above indices is that they do not take into consideration the different levels of production in each sector of the economy, what it is done by the pure linkage approach presented in the next section.

(6)

3.2. The Pure Linkage Approach

As presented by Guilhoto, Sonis and Hewings (1996) the pure linkage approach can be used to measure the importance of the sectors in terms of production generation in the economy.

Consider a two-region input-output system represented by the following block matrix, A, of direct inputs:

A A A

A A

jj jr

rj rr



F

H G I

K J

⁽³⁾

where A_jj and A_rr are the quadrate matrices of direct inputs within the first and second region and A_jr and A_rj are the rectangular matrices showing the direct inputs purchased by the second region and vice versa.

From (3), one can generate the following expression:

B I A

^B ^B

B B

I A

A I

jj jr

rj rr

jj rr

j r

jr r

rj j

 

^

 F 

H G I

K J F H G I

K JF H G I

K JF H G I K J

( )

¹ ⁰

0

0 0



 (4)

^¹

Y

⁽⁹⁾

(7)

and using the information contained in equations (4) through (8), one can derive a set of indexes that can be used: a) to rank the regions in terms of its importance in the economy; b) to see how the production process occurs in the economy.

From equations (4) and (9) one obtains:

X X

I A

A I

Y Y

j r

jj rr

j r

jr r

rj j

j r

F H

GIK J F H G I

K JF H G I

K JF H G I K JF H GIK J



^



 0

0

0 (13)

which leads to the definitions for the Pure Backward Linkage (PBL) and for the Pure Forward Linkage (PFL), i.e.,

PBL A Y

PFL A Y

r rj j j

j jr r r



 

  (14)

where the PBL will give the pure impact on the rest of the economy of the value of the total production in region j,

d i

_jY_j : i.e., the impact that is free from a) the demand inputs that region j makes from region j , and b) the feedbacks from the rest of the economy to region j and vice- versa. The PFL will give the pure impact on region j of the total production in the rest of the economy

b g

_{r r}Y ^.

3.3 The Structure of Production: Economic Landscapes

The view that has been proposed by Sonis, Hewings and Guo (1997) and by Sonis and Hewings (1999) for the interactions among the sectors to be arranged in a normalized hierarchical fashion and presented in a three-dimensional matrix that has been termed an economic landscape. This approach provides a consistent and complementary exploration of structure to the more traditional approach associated with Rasmussen and Hirschman. However, in this case, attention is directed to a matrix derived from the product of row and column multipliers extracted from the Leontief inverse matrix. This matrix, the input-output multiplier product matrix (MPM), reveals the hierarchy of backward and forward linkages and their associated economic landscapes, reflecting the cross-structure of the multiplier product matrix. The developments will be elaborated below.

(8)

Consider the column and row multipliers of the Leontief inverse defined in section 3.1 above and define b_ as been the sum over all elements of the Leontief inverse matrix.

Then, the intensity matrix, or the input-output multiplier product matrix (MPM) is defined as:

M b b b b

b b

b

b b b m

i j

n

n ij

 

F H G G G G I K J J

J J

^



 





  

1 1

1 2

1 ... 2

...

b g

^. ⁽¹⁵⁾

One can reorganize the locations of rows and columns of M in such a way that the centers of the corresponding crosses appear on the main diagonal. In this fashion, the matrix will be reorganized in such a way that a descending economic landscape will be apparent, based on the rank-size sequence of the column and row multipliers. One can reorganize the locations of rows and columns in such a way that a descending economic landscape can be apparent. Furthermore, by adopting the rank-size ordering from one economy as the numeraire, the economic landscapes can be compared visually; deviations from the smoothly descending landscape of the numeraire economy will reflect differences in economic structure. These differences will reflect variations in the industry mix of regions, variations in the degrees of intraregional intermediation as well as variations in technology.

One of the attractive features of the economic landscape analysis is that the patterns revealed are consistent with the key sector identification procedures associated with Hirschman-Rasmussen.

As Sonis, Hewings and Guo (1997) and Sonis and Hewings (1999) have pointed out, the rank- size hierarchies of the Rasmussen/Hirschman indices coincide with the rank-size hierarchies of column and row multipliers of the MPM. This rearrangement also reveals the descending rank- size hierarchies of the Rasmussen/Hirschman forward and backward linkage indices.

Thus, the economic landscape provides a complementary tool in the preliminary elaboration of differences and similarities across economies. It will not replace other techniques but will serve as a first-stage filter that may help in focusing attention on potentially important similarities and differences across economies.

(9)

3.4 The Field of Influence

The concept of field of influence was introduced and elaborated by Sonis and Hewings (1989, 1994). It is mainly concerned with the problem of coefficient change, namely the influence of a change in one or more direct coefficients on the associated Leontief inverse matrix.³ Since, given an economic system, some coefficients are more “influential” than others, the sectors responsible for the greater changes in the economy can be determined. Together with the Rasmussen/Hirschman linkage indices and the pure linkage indices, it completes our analytical framework for the determination of key sectors in an economic system.

Considering a small enough variation, , in the input coefficient, aij, the presentation of the basic solution of the coefficient change problem proposed by Sonis and Hewings may be presented as follows. let A = (aij) be an nxn matrix of direct input coefficients; let E(eij) be a matrix of incremental changes in the direct input coefficients; let B 

b g

I A ^¹  b_ij ^,

B E

bgb

^{  }I A E

g

^1^ b e( ) be the Leontief inverses before and after changes. _ij

Using the notion of inverse-important input coefficients that is based on the conception of the field of influence associated with the change in only one input coefficient, assume that this change occurs in location

b g

i j₁, ₁ , that is,

e e i i j j

i i j j

ij   

 

R S T

,

¹ or ¹

1 1

0 (16)

then, the field of influence can be constructed as the matrix F e

di

_ij generated by multiplication of the j^th column of the Leontief matrix, B, with the i^th row:

F e b b

b

b b b

ij

j j

nj

i i in

di

^

b g F

H G G G G I K J J J J

1 2

  , (17)

(10)

where F e

di

_ij ^{is a}^nxnmatrix, interpreted as the field of influence of the change on the input coefficient, a_ij. For every coefficient, a_ij, there will be an associated nxn field of influence matrix.

In order to determine which coefficients have the greater field of influence, reference is made to the rank-size ordering of the elements, S_ij , from the largest to the smallest ones. Therefore, for every matrix F e

di

_ij , there will be an associated value given by:

S_ij f_kl e_ij

l n

k





n

 di

^. ⁽¹⁸⁾

It is possible to see that S_ij b b__{j i}_ and thus provides a direct relationship with the intensity matrix defined in (15). Thus, from the values of S_ij , a hierarchy can be developed of the direct coefficients, based on their fields of influence, i.e., ranking sectoral relations in terms of their sensitivity to changes, in a sense that they will be responsible for more significant impacts on the economy. It is important to stress that each field of influence and the MPM matrix as well have a cross structure; the largest elements define the largest column and row. After exclusion of these entries, the next largest element defines the second largest cross and so on. This property is of importance in the empirical analysis.

3We considered here only the simplest case, i.e., the case in which the change occurs in only one input parameter.

However, the analysis can be extended to the cases of changes in whole rows or columns.

(11)

4. Analysis of the results

4.1 The Rasmussen/Hirschman Linkages

The results for the Rasmussen/Hirschman linkages are presented in Table 4.1 and in Figures 4.1 to 4.4 for the 26 sectors of the interregional system. In the system as a whole, the sectors that present the greater values for the backward linkages are: a) Metal Products (4), Food Products (16), and Transport Equipment (7) in the Minas Gerais State; b) Food Products (16) in the state of Pernambuco; and c) Food Products (16) in the Bahia State. In relation to the forward linkages the sectors are: a) Chemicals (11) in Bahia; b) Metal Products (4), Chemicals (11), Agriculture (1), and Trade (20) in Minas Gerais.

From Figure 4.4 one can observe that the Minas Gerais State is the one that presents the more complex productive structure of the three states considered in the analysis.

Following the definition of key sector as the one that presents the values of the backward and forward linkages greater than one (McGilvray, 1977) one has the following sectors: a) Nonmetallic Minerals (3), Metal Products (4), Machinery (5), Transport Equipment (7), Chemicals (11), Textiles (14), and Food Products (16) in Minas Gerais; b) Metal Products (4), and Chemicals (11), in Bahia; and c) Metal Products (4), Paper and Printing (9), and Food Products (16) in Pernambuco.

(12)

Table 4.1 - Rasmussen-Hirschman Linkages for São Francisco Interregional System, 1995

States

Sectors BL RBL BL RBL BL RBL FL RFL FL RFL FL RFL

1 Agriculture 0.972 43 0.950 50 0.909 63 1.636 3 1.404 8 1.323 13

2 Mining 1.099 17 1.023 30 1.010 36 0.915 32 0.968 27 0.725 63

3 Nonmetallic Minerals 1.190 7 1.021 32 1.023 29 1.062 24 0.761 54 0.832 41

4 Metal Products 1.409 0 1.153 9 1.129 12 2.602 1 1.289 15 1.238 17

5 Machinery 1.061 22 0.917 60 0.923 58 1.040 25 0.866 37 0.845 38

6 Electrical Equipament 1.063 21 0.938 55 0.994 40 0.726 62 0.671 70 0.779 50

7 Transport Equipament 1.301 2 0.962 47 0.964 45 1.089 22 0.654 75 0.672 69

8 Wood and Wood products 1.150 10 1.103 15 1.065 20 0.802 45 0.761 55 0.734 60

9 Paper and Printing 1.088 18 1.030 27 1.129 11 0.889 35 0.757 56 1.022 26

10 Rubber Products 0.991 41 1.039 25 0.903 64 0.719 64 0.757 57 0.681 67

11 Chemical Products 1.017 33 1.102 16 0.939 54 2.236 2 2.889 0 1.301 14

12 Pharmaceuticals and Cosmetics 1.041 24 1.006 38 0.999 39 0.658 74 0.651 77 0.654 76

13 Plastics 1.035 26 1.115 13 0.954 49 0.701 65 0.726 61 0.791 46

14 Textiles 1.108 14 1.017 34 1.016 35 1.082 23 0.825 42 0.957 28

15 Clothing and Footwear 1.051 23 0.946 52 1.025 28 0.672 68 0.667 72 0.688 66

16 Food and Kindred Products 1.352 1 1.223 4 1.242 3 1.133 20 0.934 31 1.109 21

17 Other Industrial Products 1.079 19 1.021 31 1.008 37 0.736 58 0.663 73 0.667 71

18 Public Utilities 0.982 42 0.971 44 0.948 51 1.402 9 1.344 12 1.174 18

19 Construction 0.961 48 0.871 68 0.888 66 0.734 59 0.780 49 0.777 51

20 Trade 0.930 56 0.940 53 0.914 61 1.435 4 1.424 5 1.417 6

21 Transport 0.963 46 0.920 59 0.890 65 1.269 16 0.949 30 1.142 19

22 Communication 0.763 73 0.749 74 0.770 72 0.788 47 0.772 53 0.782 48

23 Financial Institutions 0.822 71 0.828 70 0.854 69 0.839 39 0.904 33 0.950 29

24 Public Administration 0.651 75 0.650 77 0.650 76 0.872 36 0.836 40 0.902 34

25 Realty Services 1.193 6 1.166 8 1.208 5 0.812 43 0.775 52 0.805 44

26 Other Services 0.928 57 0.877 67 0.914 62 1.363 11 1.392 10 1.408 7

Average 1.0461 0.9822 0.9718 1.0849 0.9776 0.9374

Bahia Pernambuco Minas Gerais Bahia Pernambuco Minas Gerais

(13)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 Agriculture

Mining

Nonmetallic Minerals Metal Products

Machinery

Electrical Equipament

Transport Equipament

Wood and Wood products

Paper and Printing

Rubber Products

Chemical Products Pharmaceuticals and Cosmetics Plastics

Textiles Clothing and Footwear

Food and Kindred Products Other Industrial Products

Public Utilities Construction

Trade Transport Communication Financial Institutions

Public Administration Realty Services

Other Services

BL FL

Figure 4.1 - Rasmussen-Hirschman Linkages for Minas Gerais in the São Francisco Interregional System.

0.0 0.5 1.0 1.5 2.0 2.5 3.0 Agriculture

Mining

Machinery

Paper and Printing

Rubber Products

Other Services

BL FL

Figure 4.2 - Rasmussen-Hirschman Linkages for Bahia in the São Francisco interregional system.

(14)

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 Agriculture

Mining

Machinery

Paper and Printing

Rubber Products

Other Services

BL FL

Figure 4.3 - Rasmussen-Hirschman Linkages for Pernambuco in the São Francisco interregional system

0 0.5 1 1.5

Pernambuco

Bahia Minas Gerais

Figure 4.4 - Synthesis of average standard of the Rasmussen - Hirschman linkages for São Francisco interregional system, 1995

(15)

4.2 The Field of Influence Approach

The results for the field of influence approach are presented into Figure 4-5 where the coefficients with the 125 bigger values are showed. The dominant sectors in the system are Metal Products (4) in Minas Gerais and Chemicals (11) in Bahia. Also it can be seen that the states of Minas Gerais and Bahia are the main ones in the interregional system.

1

27

53

1 Minas Gerais 27 Bahia 53 Pernambuco

Minas GeraisBahiaPernambuco

Buying sectors

Seeling sectors

Figure 4.5 -Coefficients with the largest field of influence for São Francisco interregional system.

(16)

4.3 The Pure Linkage Approach

The results for the Pure linkage approach are presented into Table 4.31. and in Figures 4.3.1 and 4.3.2.

In terms of backward linkages the most important sectors are: a) Food Products (16), Realty Services (25), Transport Equipment (7), Metal Products (4), and Other Services (26) for Minas Gerais; b) Chemicals (11), Civil Construction (19), and Realty Services (25) for Bahia; and c) Food Products (16) for Pernambuco.

In terms of forward linkages the most important sectors are: a) Agriculture (1), Metal Products (4), Chemicals (11), Trade (20), and Other Services (26) for Minas Gerais; b) Agriculture (1), Chemicals (11), and Other Services (26) for Bahia; and c) Other Services (26) for Pernambuco.

The analysis of the Pure linkages reinforce the importance of the Minas Gerais State in the interregional system as showed above.

It is important to call attention for the fact that while the sectors of Metal Products (4) in Minas Gerais and Chemicals (11) in Bahia are key sectors for the system, the Chemical sector does not use water from the São Francisco river basin while the same is not true for the Metal Products sector in Minas Gerais.

The values for the total Pure linkage show that the most important sectors are: a) Agriculture (1), Metal Products (4), Food Products (16), Trade (20), and Other Services (26) for Minas Gerais; and b) Chemicals (11) for Bahia.

(17)

Table 4.3.1 - Pure Linkages of the São Francisco Interregional System, 1995.

PBL RPBL PBL RPBL PBL RPBL PFL RPFL PFL RPFL PFL RPFL PTL RPTL PTL RPTL PTL RPTL 1 Agriculture 1933431 7 676169 21 331729 28 3559473 0 1182861 10 739602 16 5492904 2 1859030 14 1071331 26

2 Mining 1155454 14 -73775 78 14964 69 465038 26 784541 15 32875 67 1620492 18 710767 32 47839 72

3 Nonmetallic Minerals 254705 33 12265 70 5924 75 967890 13 205844 47 176171 49 1222595 23 218109 54 182095 56 4 Metal Products 2357573 3 294228 30 147051 39 3519383 1 497464 23 252182 38 5876956 0 791693 30 399233 42 5 Machinery 465662 23 127726 42 81621 46 1045536 12 266606 37 101691 56 1511198 22 394333 43 183313 55 6 Electrical Equipament 412981 24 49066 55 138795 41 161250 50 32880 66 90316 59 574230 36 81945 68 229111 53 7 Transport Equipament 2697698 2 19053 68 31162 61 719878 18 5509 78 13072 71 3417577 8 24562 75 44234 73 8 Wood and Wood products 374597 25 124854 43 50299 53 242420 42 118519 53 51335 63 617017 35 243373 51 101634 66 9 Paper and Printing 172813 38 49627 54 40325 58 486222 25 112627 55 245692 41 659035 34 162253 57 286017 48

10 Rubber Products 10924 72 23244 64 2841 76 98827 58 49684 64 10090 74 109751 64 72928 69 12932 78

11 Chemical Products 265084 32 2133447 6 56662 51 3502079 2 2002617 5 408309 30 3767163 7 4136065 5 464971 41 12 Pharmaceuticals and Cosmetics 233868 35 55411 52 43636 56 47102 65 11903 72 10406 73 280970 49 67314 70 54042 71

13 Plastics 26230 62 25641 63 9306 73 116116 54 89659 60 99688 57 142346 59 115299 61 108994 65

14 Textiles 225372 36 68392 49 35451 59 300458 35 55679 62 78289 61 525829 40 124071 60 113740 63

15 Clothing and Footwear 272042 31 81074 48 107763 44 16902 69 5835 76 6445 75 288943 47 86909 67 114209 62 16 Food and Kindred Products 4403530 0 1333803 11 1171934 13 1314318 9 418393 29 377820 31 5717848 1 1752196 16 1549753 20 17 Other Industrial Products 105746 45 11680 71 7312 74 222212 44 19806 68 14172 70 327958 44 31487 74 21485 76 18 Public Utilities 318825 29 237683 34 42508 57 1659137 7 714033 19 268463 36 1977962 13 951716 28 310971 46 19 Construction 2290149 5 1566671 9 866300 19 314232 33 245938 40 149280 51 2604381 10 1812610 15 1015579 27 20 Trade 1775268 8 1233544 12 874306 18 2499323 4 1167355 11 674477 20 4274591 4 2400899 11 1548783 21 21 Transport 1045591 15 180844 37 145411 40 1829131 6 502145 22 423293 28 2874721 9 682989 33 568705 38 22 Communication 81360 47 34377 60 22627 65 459788 27 207141 46 139264 52 541148 39 241518 52 161890 58 23 Financial Institutions 578784 22 374305 26 350460 27 536940 21 371889 32 222906 43 1115724 25 746194 31 573365 37 24 Public Administration 59280 50 22004 66 19428 67 734322 17 302284 34 251195 39 793603 29 324288 45 270623 50 25 Realty Services 3636837 1 1383112 10 943540 17 494387 24 221127 45 180553 48 4131224 6 1604238 19 1124092 24 26 Other Services 2291998 4 979073 16 785321 20 2652520 3 1337427 8 867039 14 4944518 3 2316500 12 1652359 17

Average 1055608 423981 243334 1075572 420376 226332 2131180 844357.2 469665.4

Sector Minas Gerais Bahia Pernambuco Minas Gerais Bahia Pernambuco Minas Gerais Bahia Pernambuco

-500000 0 500000 1000000 1500000 2000000 2500000 3000000 3500000 4000000 4500000 5000000

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

PBL MG PBL BA PBL PE R$1000

Figure 4.3.1 - Pure backward linkage of the São Francisco Interregional System.

(18)

0 500000 1000000 1500000 2000000 2500000 3000000 3500000 4000000

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

PFL MG PFL BA PFL PE R$1000

Figure 4.3.2 Pure forward linkage of the São Francisco Interregional System.

4.4 Intensity Matrices - Applications to the São Francisco Interregional System

The results for the intensity matrices are presented into figures 4.4.1 to 4.4.3 taking the hierarchy of Minas Gerais State as the basis for comparison. From the figures one can see that the states have different productive structure. The Minas Gerais State shows a more linked industrial structure while for the Bahia State the bigger relations occur in the Chemicals sector, the Pernambuco State shows to be the one that has a less complex productive structure with a greater importance for the service related sectors.

(19)

4 11 1 20 18 26 21 16 14 3 7 5 9 24 2 23 25 8 22 19 17 6 10 13 15 12

24 20

19 18

12 5

2 3

16

0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18 0.20

row hierarchy of forward linkages

column hierarchy of backward linkages

Figure 4.4.1- Minas Gerais: Cross-structure "landscape for first order intensity field

4 11 1 20 18 26

21 16 14 3 7 5 9 24 2 23 25 8 22 19 17 6

10 13 15 12 24

20 19

18 12

5 2

3 16

0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18 0.20

Figure 4.4.2 - Bahia: Cross-structure "landscape" using Minas Gerais imposed hierarchy

(20)

4 11 1 20 18 2621 16 14 3 7 5 9 24 2 23 25 8

22 19 17 6 10 13 15 12 24

20 19

18 12

5 2

3 16

0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18 0.20

Figure 4.4.3 - Pernambuco: Cross-structure "landscape" using Minas Gerais imposed hierarchy 4.5. Water Use in the São Francisco River Basin

As show in Table 4.5.1 the greater volume of water withdraw is made by the Bahia State with 359874 l/s, followed by the Minas Gerais State with 184700 l/s, and finally the Pernambuco State with 145145 l/s. In terms of water use by category, agriculture uses around 59% of the water withdraw, while the industrial sector uses around 20% (Table 4.5.2) .

Table 4.5.1 Total water withdrawals by water-use category, 1995

liter per second

Minas Gerais Bahia Pernambuco

Irrigation 115611.24 259258.48 75145.54

Industrial¹ 56 1342.73 -

Domestic and

commercial 69033 79273 70000

Total 184700.24 359874.21 145145.54

Note: 1) Partial withdrawal Source: SRH, 1995.

(21)

Table 4.5.2 Proportion of the water-use by category, 1995

Minas Gerais Bahia Pernambuco USA

Irrigation 0.5910 0.5910 0.5910 0.3930

Industry 0.1966 0.1966 0.1966 0.0607

Users Publics 0.1271 0.1271 0.1271 -

Commercial 0.0430 0.0430 0.0430 0.0085

Domestic 0.0430 0.0430 0.0430 0.0099

Source: IGAM, 1998; COPASA, 1999; Solley; Pierce e Perlman, 1998.

Using different sources of information and different techniques of estimation it was possible to calculate the volume of water use for the interregional system as showed into Table 4.5.3.

The sectors that use more water in the three states are: a) Agriculture (1), Mining (2), and Chemical (11), in Minas Gerais; b) Agriculture (1), Mining (2), and Chemical (11) in Bahia; and c) Agriculture (1), Mining (2), and Public Utilities (18) in Pernambuco.

From the results presented in Table 4.5.3 and the total value of production in each sector in each state it was possible to estimate the interregional freshwater withdrawal coefficients (m³/R$) as showed into Table 4.5.4.

In analyzing the results presented into Table 4.5.4 some interesting points are revealed like that fact that the coefficients of the Mining (2) sector are greater than the ones for the Agriculture (1) sector in the States of Minas Gerais and Pernambuco.

As this work is still under way the next step in process of analysis is the construction of an interregional water-content matrix such that it can be used to estimate interregional multipliers of water use allowing in this way a better understanding o how the use of water take place among the regions and the sectors being analyzed in this work.

(22)

Table 4.5.3 - Estimate water-use by sector

States and Sectors

1 Agriculture 115.611240 62.26 212.704230 59.10 85.788505 83.95

2 Mining 32.695670 17.61 28.470795 7.91 4.256235 4.16

3 Nonmetallic Minerals 7.508742 4.04 7.310330 2.03 0 0

4 Metal Products 1.934203 1.04 0.700109 0.19 0 0

5 Machinery 0.023039 0.01 0.031441 0.01 0 0

6 Electrical Equipament 0.009353 0.01 0 0 0 0

7 Transport Equipament 0.179154 0.10 0 0 0 0

8 Wood and Wood products 0.418963 0.23 0.415596 0.12 0.407000 0.40

9 Paper and Printing 0.573277 0.31 0 0 0 0

10 Rubber Products 0.003325 0.002 0 0 0 0

11 Chemical Products 11.898565 6.41 83.825047 23.29 0 0

12 Pharmaceuticals and Cosmetics 1.136132 0.61 1.247520 0.35 0 0

13 Plastics 0.423936 0.23 0 0 0 0

14 Textiles 0.006077 0.003 0.009291 0.003 0.007492 0.01

15 Clothing and Footwear 0.011339 0.01 0.001451 0.0004 0.029821 0.03

16 Food and Kindred Products 1.061120 0.57 1.426871 0.40 2.016635 1.97

17 Other Industrial Products 0.005905 0.003 0.000960 0.0003 0.002424 0.002

18 Public Utilities 7.872712 4.24 15.339374 4.26 6.186722 6.05

19 Construction 0.040075 0.02 0.065528 0.02 0.124612 0.12

20 Trade 0.701502 0.38 1.366822 0.38 0.551271 0.54

21 Transport 0.486805 0.26 0.458958 0.13 0.233573 0.23

22 Communication 0.098808 0.05 0.198008 0.06 0.065679 0.06

23 Financial Institutions 0.341815 0.18 1.006992 0.28 0.448931 0.44

24 Public Administration 1.196041 0.64 2.316157 0.64 0.964097 0.94

25 Realty Services 0.539285 0.29 0.930456 0.26 0.334831 0.33

26 Other Services 0.928242 0.50 2.086206 0.58 0.774851 0.76

Total 185.705323 100.00 359.912142 100.00 102.192676 100.00

Minas Gerais Bahia Pernambuco

Water withdrawal

(m³)

Water withdrawal

(m³)

Water withdrawal

(m³)

% % %

(23)

Table 4.5.4 Interregional freshwater withdrawal coefficients

States and Minas Gerais Bahia Pernambuco

Sectors m³ /R$ m³ /R$ m³ /R$

1 Agriculture 0.010367 0.058147 0.040100

2 Mining 0.013618 0.043420 0.070669

3 Nonmetallic Minerals 0.004926 0.031416 0

4 Metal Products 0.000170 0.000543 0

5 Machinery 0.000012 0.000052 0

6 Electrical Equipament 0.000011 0 0

7 Transport Equipament 0.000033 0 0

8 Wood and Wood products 0.000471 0.001202 0.002775

9 Paper and Printing 0.000680 0 0

10 Rubber Products 0.000027 0 0

11 Chemical Products 0.002626 0.008374 0

12 Pharmaceuticals and Cosmetics 0.002626 0.011165 0

13 Plastics 0.002626 0 0

14 Textiles 0.000007 0.000042 0.000038

15 Clothing and Footwear 0.000023 0.000007 0.000138

16 Food and Kindred Products 0.000134 0.000570 0.000789

17 Other Industrial Products 0.000015 0.000024 0.000088

18 Public Utilities 0.001920 0.005457 0.010903

19 Construction 0.000007 0.000012 0.000043

20 Trade 0.000099 0.000321 0.000177

21 Transport 0.000108 0.000462 0.000253

22 Communication 0.000108 0.000462 0.000253

23 Financial Institutions 0.000108 0.000462 0.000253

24 Public Administration 0.000108 0.000462 0.000253

25 Realty Services 0.000108 0.000462 0.000253

26 Other Services 0.000108 0.000462 0.000253

(24)

5. Conclusions

The motivation for the work done here was the need for a model that could be used to study the interrelations among the regions and the sectors in respect to water use in the São Francisco river basin.

As a starting point, from the national input-output tables for the Brazilian economy for the year of 1995 and using information from various census it was constructed an interregional system comprising the 3 major states in the São Francisco river basin, i.e., Minas Gerais, Bahia, and Pernambuco.

The analysis of the interregional system has show that the more complex productive structure is found in the Minas Gerais State which has the Metal Products (4) as the main sector in the system. In second place comes the state of Bahia in which the main sector is the Chemical (11). The less complex productive structure is found in the Pernambuco State.

In terms of water use one has that the in terms of volume the sectors that use more of this resource are: a) Agriculture (1), Mining (2), and Chemical (11), in Minas Gerais; b) Agriculture (1), Mining (2), and Chemical (11) in Bahia; and c) Agriculture (1), Mining (2), and Public Utilities (18) in Pernambuco.

In term of freshwater withdrawal coefficients (m³/R$) some interesting points are revealed like the fact that the coefficients of the Mining (2) sector are greater than the ones for the Agriculture (1) sector in the States of Minas Gerais and Pernambuco.

As this work is still under way the next step in process of analysis is the construction of an interregional water-content matrix such that it can be used to estimate interregional multipliers of water use, allowing in this way to a better understanding of how the use of water take place among the regions and the sectors being analyzed in this work.

(25)

6. References

ALVES, Francisco Ferreira; GOMES, Francisco de Assis L.; SILVA FILHO, Guerino E. da;

ARY, José Carlos Azyz; VIANA, Manuel O. de Lima e SOARES, Paulo Henrique. O setor industrial do Nordeste: diagnóstico, modernização e concentração espacial. Diretrizes para um plano de ação do BNB (1991-1995); Setor Secundário. Fortaleza, 1997. 7v., v.3. p.9 - 260. (BNB. Estudos Econômicos e Sociais, v.59).

BDMG (Banco de Desenvolvimento de Minas Gerais). Economia mineira 1989: diagnósticos e perspectivas. Belo Horizonte, 1989. 5 v. em 7.

BDMG (Banco de Desenvolvimento de Minas Gerais). Economia mineira 1989: diagnósticos e perspectivas. Belo Horizonte, 1989a. v.2: Indústria. t.1. Novos Desafios. 114 p.

CODEVASF - Companhia de Desenvolvimento do Vale do São Francisco. Projeto semi-árido:

proposta de desenvolvimento sustentável da bacia do rio São Francisco e do semi-árido nordestino. Brasília: CODEVASF, 1996. 60p.

_________ . Produção agrícola de 1997 / Companhia de Desenvolvimento do Vale do São Francisco. Brasília: CODEVASF. Diretoria de Planejamento. Coordenadoria de Acompanhamento e Avaliação. 1998a. 71p.

_________ . Relatório anual de monitoria. Diretoria de Planejamento. Brasília: CODEVASF.

1995;1996b.

_________. O Vale do São Francisco. (http://www.codevasf.gov.br.),1998b.

CONSIDERA, Cláudio M. e MEDINA, Mérida H. PIB por unidade da Federação: Valores correntes e constantes - 1985/96. Brasília, DF. IPEA, 1998 (Texto para Discussão n.^o 610).

32p. + 1 disquete.

COPASA MG . Companhia de Saneamento de Minas Gerais - COPASA - MG. Belo Horizonte, 1999. (Relatório de computador).

EMBRAPA. Sistema de monitoramento por satélite. (http//www.nma.embrapa.br), 1999.

FERNANDES, C. L. de L. A inserção de Minas Gerais na economia nacional: uma análise de insumo-produto inter-regional. Rio de Janeiro, 1997. Tese (Doutorado). Universidade Federal do Rio de Janeiro.

GUILHOTO, Joaquim J. Martins; SONIS, Michael; HEWINGS, Geoffrey John Dennis e MARTINS, Eduardo Borges. Índices de ligações e setores-chave na economia brasileira:

1959/80. Pesquisa e Planejamento Econômico. Rio de Janeiro, v. 24, n. 2, p. 287-314, ago.

1994.

GUILHOTO, Joaquim J. Martins; SONIS, Michael e HEWINGS, Geoffrey John Dennis.Linkages and multipliers in a multirregional framework: integration of alternative approaches. REAL 96-T-8, nov., 1996. 20p.

(26)

GUILHOTO, Joaquim J. Martins; HEWINGS, Geoffrey John Dennis e SONIS, Michael.

Interdependence, linkages and multipliers in Asia: an international input-output analysis. REAL 97-T-2, oct., 1997, 33p.

HEWINGS, Geoffrey John Dennis; FONSECA, Manuel; GUILHOTO, Joaquim J. Martins e SONIS, Michael. Key sectors and structural change in the brasilian economy: a comparison of alternative approaches and their policy implications. Journal of Policy Modeling, v.11, n.1, 1989, p. 67-90.

IBGE (Instituto Brasileiro de Geografia e Estatística). Anuário Estatístico do Brasil. 1996, 1997.

IGAM (Instituto Mineiro de Gestão das Águas). Listagem Rio São Francisco. Belo Horizonte, 1998. + 1 disquete.

LAVINAS, Lena; GARCIA, Eduardo Henrique e AMARAL, Rubens Marcelo do.

Desigualdades regionais e retomada do crescimento num quadro de integração econômica. Instituto de Pesquisa Econômica Aplicada (IPEA), Rio de Janeiro (RJ), mar.

1997. 37 p. (Texto para Discussão n. 466).

MELO, Ademir Alves de e DUARTE, Renato Santos. A dinâmica da economia do Nordeste nas décadas de 1970 e 1980 e perspectivas para os anos 1990. In.: Diretrizes para um plano de ação do BNB (1991-1995); Infra-estrutura econômica e social do Nordeste. Fortaleza, 1997. 7v., v.6, t 2. p. 32-73. (BNB. Estudos Econômicos e Sociais, v.59).

RURALMINAS (Fundação Rural Mineira - Colonização e Desenvolvimento Agrário).

Inventário dos recursos hídricos regionais: levantamento de usuários (Tomo XI).

Consórcio ECOPLAN-MACNA-CAB. (Relatório Preliminar). Junho, 1999.

SETTI, Arnaldo Augusto. Legislação para uso dos recursos hídricos. Brasília, DF: ABEAS;

Viçosa, MG: UFV; Departamento de Engenharia Agrícola, 1998. 227p.

SOLLEY, Wayne B.; PIERCE, Robert R. e PERLMAN, Howard A. Estimated Use of Water in the United States in 1995. U.S. Geological Survey (USGS) Circular 1200. Denver, 1998.

71p.

SONIS, Michael e HEWINGS, Geoffrey J. D. Error and Sensitivity input-output analysis: a new approach. 1989.

SONIS, M.; HEWINGS, G. J. D. "Economic Landscapes: Multiplier Product Matrix Analysis for Multiregional Input-Output Systems," Hitotsubashi Journal of Economics 40(1), pp. 59-74.

1999.

SONIS, M.; HEWINGS, G. J. D; GUO, J. “Input-output multiplier product matrix.” Discussion Paper 94-T-12 (revised, 1997), Regional Economics Applications Laboratory, University of Illinois. 1997

SONIS, Michael; GUILHOTO, Joaquim J. Martins; HEWINGS, Geoffrey J. D. e MARTINS, Eduardo B. Linkages, key sectors, and structural change: some new perspectives. The Developing Economies, v. XXXIII, n.3, sep., 1995. p.233-269.

(27)

SRH (Secretaria de Recursos Hídricos). Ministério do Meio ambiente, dos Recursos Hídricos e da Amazônia Legal. Brasília (DF),1999. (Relatório de computador).